Abstract Lacunar strokes are small infarctions of deep penetrator end arteries that supply deep areas of brain including the basal ganglia, internal capsule, thalamus and pons. The ?cause? of lacunar strokes has been controversial. C Miller Fisher was the first to suggest that hypertension was one cause, noting lipohyalinosis (inflammation, fibrinoid necrosis of vessel wall) in smaller penetrating arteries (<200um) occurred only in hypertensives. He also showed, however, that larger lacunar strokes were often associated with microatheroma associated narrowing of larger penetrating arteries. This pathology is now thought to be the most common and associated with lipid abnormalities and diabetes. Moreover, some lacunar strokes are now thought to be associated with intracranial atherosclerosis of the parent artery at the origin of the penetrators. Though lacunar strokes were not traditionally associated with large vessel extracranial atherosclerosis or cardioembolism, there are now a number of cases where lacunar strokes have been associated with large vessel atherosclerosis or cardiac causes. Indeed, in our own recent study we showed that as many as ~50% of small deep infarcts (>15mm in size) were predicted to be due to large vessel atherosclerosis or to cardioembolic disease. It is crucial to identify the correct causes of lacunar strokes because treatments differ: drugs for hypertension vs drugs for lipid abnormalities, surgery for large vessel carotid disease, anticoagulants for cardioembolic disease, and anti-platelet agents for lacunar strokes associated with hypertension or abnormal lipids or diabetes. The biology of lacunar strokes has been studied very little, and there are no accepted biomarkers for lacunar stroke or its causes. Our group, however, has begun to provide novel insights into how the immune and clotting profiles in peripheral blood differ in lacunar strokes compared to large vessel and cardioembolic strokes. We have shown that the immune response in large vessel and cardioembolic cortical strokes is mainly associated with neutrophil genes, whereas the immune response in lacunar stroke associated with hypertension /lipids/ diabetes is associated with inflammatory monocyte-related genes. However, since the preliminary studies were performed on whole blood, there is a great need to study individual immune cell types in these different causes of ischemic stroke. Therefore, in this study we will isolate neutrophils and monocytes in patients with lacunar stroke compared to large vessel and cardioembolic stroke. We will perform RNA sequencing on those cells in the following aims which allows us to measure expression levels of >250,000 alternatively spliced transcripts which are derived from the ~20,000 coding genes. Specific Aim #1. Use RNAseq to demonstrate that alternatively spliced transcript level expression profiles in monocytes, neutrophils and whole blood differ for subcortical lacunar strokes associated with hypertension, abnormal lipids, diabetes, large vessel atherosclerosis, or cardioembolism compared to each other and compared to matched controls. Specific Aim #2. Use RNAseq to demonstrate that alternatively spliced transcript level expression profiles for subcortical lacunar strokes with hypertension, lipids or diabetes are associated with monocytes, and differ from cortical large vessel and cardioembolic strokes which are associated mainly with neutrophils. Specific Aim #3a. Use Support Vector Machine (SVM) to derive the minimum number of transcripts from Aim #1 that differentiate the different causes of lacunar strokes including hypertension, abnormal lipids, diabetes, large vessel atherosclerosis, or cardiac disease. Specific Aim #3b. Use the profiles derived in Aim #3a to predict the cause of lacunar stroke in a second cohort using qRT-PCR to measure expression and SVM to perform predictions with >85% sensitivity and specificity. Significance: Identifying the causes of lacunar strokes based upon splice variant expression in individual blood cell types is important since it would affect therapy used for secondary stroke prevention.